ERG in Ophthalmology: Normal Waves and How to Read Them
This article explains how to interpret electroretinography (ERG).
When interpreting ERG results, it is crucial to recall the anatomy of the retina. Instead of simply memorizing the findings, understanding which part of the retina is impaired and how that translates into specific ERG changes will enable a much deeper level of interpretation.
In this article—and in all educational content I produce—I aim to explain not only the facts but the underlying logic and physiology behind each finding. I hope this helps lighten the cognitive load in your exam preparation and deepens your clinical understanding.
Prefer to watch? This 16-minute video walks through the same logic, from the normal ERG waveforms to the three causes of an electronegative (negative) ERG.
How Light and Signals Travel Through the Retina
Before we look at any waveform, there is one thing to understand first: how light actually reaches the photoreceptors.
The retina is a transparent tissue. When light enters the eye, it does not stop at the surface of the retina. It passes straight through and reaches the photoreceptors, which sit in the deepest layer.
There, the photoreceptors are stimulated, and the signal travels back toward the surface: from the photoreceptors to the bipolar cells, then to the ganglion cells, and finally out through the optic nerve.
Light goes down, and the signal comes back up. Everything in this article is built on this one idea.
Overview of Normal ERG Waveforms
Before diving into the interpretation of ERG findings, let’s first review the typical waveforms seen in a normal electroretinogram (ERG). Understanding what is considered normal provides a critical foundation for recognizing pathological changes.
ERG measures the electrical responses of the retina to light. Its strength is that it can detect functional abnormalities even when the retina looks normal on examination.
Rod response
- Stimulus: Dim light under dark-adapted conditions
- Key Feature: Small b-wave only (no a-wave)
- Clinical Note: Isolates rod function; used to assess scotopic vision
Mixed response
- Stimulus: Bright flash under dark adaptation
- Key Features: Large a-wave followed by b-wave; includes oscillatory potentials (OPs)
- Clinical Note: Reflects both rod and cone function; a-wave = photoreceptors, OPs = amacrine cells, b-wave = bipolar cells
In the international (ISCEV) standard, this response is the dark-adapted 3.0 ERG. It is also called the combined rod-cone response. In this article we call it the mixed response.
Cone response
- Stimulus: Bright flash under light-adapted conditions
- Key Feature: Fast, lower amplitude a- and b-waves
- Clinical Note: Isolates cone function; used to assess photopic vision
Flicker response
- Stimulus: 30 Hz flickering light
- Key Feature: Regular, sinusoidal waveform with no a-wave
- Clinical Note: Reflects cone function; rods cannot follow high-frequency flicker
All four responses matter. But one of them is especially important for telling retinal diseases apart, and that is the mixed response. The rest of this article focuses on it.
The Mixed Response on the Flash ERG
The mixed response is the most important component of the ERG examination. The schematic above provides a simple overview of the retina and shows how each retinal cell type corresponds to specific components of its waveform.
The initial downward deflection is the a-wave, followed by small jagged waves known as oscillatory potentials (OPs), and then a prominent upward deflection called the b-wave.
- The a-wave reflects activity from photoreceptors (rods and cones).
- The OPs reflect the function of amacrine cells.
- The b-wave corresponds to bipolar cell function (with a contribution from Müller cells).
Even if you cannot remember every detail, keep this simple mapping in mind: the a-wave comes from the outer retina, and the OPs and the b-wave come from the inner retina. This mapping is the key to everything that follows.
Abnormal Findings in Flash ERG
Abnormal flash ERG findings can be broadly categorized into three types. We will start with the two easier ones and save the hardest, the negative ERG, for last.
1. Loss of OPs
Loss of oscillatory potentials is commonly seen in conditions such as diabetic retinopathy (DR) and retinal vein occlusion (RVO). The a-wave and the b-wave are still present, but the small wavelets on the rising b-wave are reduced or gone.
Recall that OPs are generated by amacrine cells, which reside in the inner retinal layers. The inner retina is supplied by the retinal circulation. Because vascular diseases tend to affect the retina from the inner layers outward, OPs are often the first component to disappear in these conditions.
In short, reduced OPs tell you that the inner retina is ischemic, in its early or milder form.
2. Flat ERG
In a flat ERG, almost nothing can be recorded: no a-wave, no OPs, no b-wave.
The problem is in the photoreceptors themselves. If the photoreceptors do not respond, nothing downstream responds either. No signal reaches the bipolar cells, so there is no b-wave and no OPs. The whole waveform disappears.
The classic example is retinitis pigmentosa (RP), where both rod and cone photoreceptors degenerate across the whole retina. Other end-stage retinal degenerations can show a flat ERG as well.
3. Negative ERG (Electronegative ERG)
A negative ERG, also called an electronegative ERG, is a mixed response in which the b-wave amplitude is smaller than the a-wave. The waveform stays below the baseline and looks “negative.”
What does it mean? The photoreceptors are working, so the a-wave is preserved. But the signal does not get through to the bipolar cells, so the b-wave is lost.
This is the pattern most people find confusing, because it does not have one cause. It has three.
- The bipolar cells themselves are not working.
- The bipolar cells are alive, but the signal from the photoreceptors cannot reach them.
- Dark adaptation was insufficient.
Negative ERG, Cause 1: Bipolar Cell Dysfunction
Here, the bipolar cells themselves are not working, so the b-wave cannot rise. This happens in two ways.
The first is a disease of the bipolar cells. The classic example is congenital stationary night blindness (CSNB).
The second is inner retinal ischemia. Note the difference from the loss-of-OPs pattern above. With milder ischemia, only the amacrine cells drop out, and you lose the OPs. With severe ischemia, the bipolar cells are lost too, and the waveform becomes negative. Advanced DR and extensive RVO can reach this stage.
Special Case: CRAO
Let’s consider the ERG findings in central retinal artery occlusion (CRAO).
Among the 10 layers of the retina, the inner layers (from the internal limiting membrane to the inner nuclear layer) are supplied by the retinal circulation, whereas the outer layers (from the outer plexiform layer to the retinal pigment epithelium) receive nutrients from the choroidal circulation.
CRAO cuts off the retinal circulation, so all of these inner layers become ischemic at once, including the inner nuclear layer where the bipolar cells reside.
Therefore, in CRAO, the ERG loses its OPs and becomes negative at the same time.
On the left, a cherry-red spot at the fovea stands out against the pale retina. On the right, the b-wave of the right eye (R) does not climb back to the baseline: a negative ERG.
Many types of uveitis, such as sarcoidosis, mainly inflame the retinal veins (phlebitis). Bacterial endophthalmitis is different: it inflames the arteries as well as the veins. As a result, severe cases of infectious endophthalmitis can resemble CRAO and produce a negative ERG.
Imagine a patient with bacterial endophthalmitis whose vitreous is so opaque that you cannot see the fundus at all. If you record the ERG and the mixed response is negative, that tells you the inflammation has already reached the retina. It is a sign of deep damage, and it indicates the urgency of surgical intervention.
Negative ERG, Cause 2: The Signal Cannot Reach the Bipolar Cells
In this case, the photoreceptors work and the bipolar cells are alive, but the connection between them is broken. The a-wave is preserved, the b-wave is lost, and the ERG is negative.
The representative disease is X-linked retinoschisis (XLRS). In retinoschisis, the retina splits within its inner layers. On OCT, the split is most often seen in the inner nuclear layer.
The underlying problem is a defective protein called retinoschisin, which normally holds the retinal cells together, including the synapses between the photoreceptors and the bipolar cells. So the photoreceptors respond, but their signal cannot get across to the bipolar cells.
Negative ERG, Cause 3: Insufficient Dark Adaptation
Before measuring rod responses with ERG, patients must undergo proper dark adaptation.
In this test, a dim light stimulus—too weak to stimulate cones—is used to isolate rod function.
Let’s consider how the ERG waveform changes as light intensity is gradually increased from a very dim level.
At low light levels, a typical rod response is observed. As the intensity increases, the a-wave begins to appear, cutting downward into the waveform. With further increases, the waveform progresses into the full mixed response.
The mixed response reflects the combined activity of both rod and cone photoreceptors, depending on the light intensity used during stimulation.
However, if proper dark adaptation is not performed, even a normal subject may show a negative ERG. To understand why, we need to go one step deeper into the b-wave.
ON and OFF bipolar cells
More precisely, the b-wave of the mixed response comes mainly from the ON bipolar cells. Bipolar cells come in two types, ON and OFF, and here is the key point:
- Rods connect only to ON bipolar cells.
- Cones connect to both ON and OFF bipolar cells.
Under proper dark adaptation, both rods and cones respond to the flash. The rod signal goes to ON bipolar cells, part of the cone signal also goes to ON bipolar cells, and together they produce a large b-wave.
What happens when dark adaptation is insufficient
If dark adaptation is insufficient, the rods cannot respond yet, so the ON bipolar cells connected to the rods stay silent. The cones do respond, but their signal is split between ON and OFF bipolar cells.
So the ON bipolar response, and therefore the b-wave, becomes small compared with the a-wave from the cones. The result is a negative ERG.
Oguchi disease and fundus albipunctatus
In diseases like Oguchi disease or fundus albipunctatus, dark adaptation is abnormally delayed, and even after standard adaptation, the ERG may still appear negative due to insufficient rod response.
In a healthy eye, 20 minutes of dark adaptation is enough for the rods to respond. In these two diseases, the rods need much longer. If you record the ERG after the routine 20 minutes, the rods are not ready, and you get a negative waveform. In fundus albipunctatus, two to three hours of dark adaptation brings the ERG back to normal or near normal. In Oguchi disease, the rods take even longer, and the ERG may stay smaller than normal.
It is the same mechanism as insufficient dark adaptation. Only the cause is in the patient, not in the protocol.
Light Adaptation
The cone response and flicker response mainly reflect cone photoreceptor function, as rods cannot follow high-frequency stimuli like flicker.
Therefore, in diseases that selectively affect cones—such as cone dystrophy or achromatopsia—the cone and flicker responses are almost completely absent.
Summary
- Link each wave with its cells: the a-wave with the photoreceptors, the OPs with the amacrine cells, and the b-wave with the bipolar cells.
- Know the three abnormal patterns of the mixed response: loss of OPs, the flat ERG, and the negative (electronegative) ERG.
- When you see a negative ERG, think of the three causes: bipolar cell dysfunction, blocked signaling from the photoreceptors to the bipolar cells, and insufficient dark adaptation.
Which retinal layer makes which wave, the three abnormal patterns, and the dark-adaptation trap to check before you report “negative” — on one page, by email. One email when I publish something new, about once a month. Unsubscribe anytime.
